WO2019097335A1 - Phonetic patterns for fuzzy matching in natural language processing - Google Patents
Phonetic patterns for fuzzy matching in natural language processing Download PDFInfo
- Publication number
- WO2019097335A1 WO2019097335A1 PCT/IB2018/058521 IB2018058521W WO2019097335A1 WO 2019097335 A1 WO2019097335 A1 WO 2019097335A1 IB 2018058521 W IB2018058521 W IB 2018058521W WO 2019097335 A1 WO2019097335 A1 WO 2019097335A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- token
- phonetic
- pattern
- phonetic pattern
- computer
- Prior art date
Links
- 238000003058 natural language processing Methods 0.000 title description 23
- 230000015654 memory Effects 0.000 claims abstract description 35
- 239000011159 matrix material Substances 0.000 claims abstract description 23
- 238000013500 data storage Methods 0.000 claims abstract description 6
- 238000012545 processing Methods 0.000 claims description 80
- 238000000034 method Methods 0.000 claims description 54
- 238000010586 diagram Methods 0.000 description 19
- 230000008569 process Effects 0.000 description 19
- 230000006870 function Effects 0.000 description 10
- 238000004590 computer program Methods 0.000 description 8
- 241000227653 Lycopersicon Species 0.000 description 7
- 235000007688 Lycopersicon esculentum Nutrition 0.000 description 7
- 238000004891 communication Methods 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 5
- 241000282412 Homo Species 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000003491 array Methods 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 239000004744 fabric Substances 0.000 description 2
- 239000000835 fiber Substances 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000001902 propagating effect Effects 0.000 description 2
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 1
- 244000061456 Solanum tuberosum Species 0.000 description 1
- 235000002595 Solanum tuberosum Nutrition 0.000 description 1
- 230000001427 coherent effect Effects 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 229910052802 copper Inorganic materials 0.000 description 1
- 239000010949 copper Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000003014 reinforcing effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000013518 transcription Methods 0.000 description 1
- 230000035897 transcription Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 230000003313 weakening effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/08—Speech classification or search
- G10L15/18—Speech classification or search using natural language modelling
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/08—Speech classification or search
- G10L15/18—Speech classification or search using natural language modelling
- G10L15/183—Speech classification or search using natural language modelling using context dependencies, e.g. language models
- G10L15/187—Phonemic context, e.g. pronunciation rules, phonotactical constraints or phoneme n-grams
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/02—Feature extraction for speech recognition; Selection of recognition unit
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/02—Feature extraction for speech recognition; Selection of recognition unit
- G10L2015/027—Syllables being the recognition units
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/27—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique
- G10L25/33—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique using fuzzy logic
Definitions
- the present invention relates generally to Natural Language Processing (NLP) and more particularly to a method, system, and computer program product for phonetic patterns for fuzzy matching in natural language processing.
- NLP Natural Language Processing
- a natural language is a scripted (written) or a vocalized (spoken) language having a form that is employed by humans for primarily communicating with other humans or with systems having a natural language interface.
- Natural language processing is a technique that facilitates exchange of information between humans and data processing systems.
- one branch of NLP pertains to transforming human readable or human understandable content into machine usable data.
- NLP engines are presently usable to accept input content such as a newspaper article or human speech, and produce structured data, such as an outline of the input content, most significant and least significant parts, a subject, a reference, dependencies within the content, and the like, from the given content.
- An NL input is an input constructed using a grammar of a natural language and presented in a suitable form, including but not limited to text, audio, and forms thereof, such as transcription from audio speech, machine-generated audio from text.
- a unit of an NL input is the shortest meaningful portion of the input. For example, in the English language, a unit would be a word; and words form other larger structures such as phrases, sentences, and paragraphs in the NL input.
- a unit of an NL input is also referred to herein as a token.
- the illustrative embodiments recognize that machine-understanding of a token is sensitive to a number of factors. In some cases, an emphasis placed on a token or a portion thereof can cause an incorrect selection corresponding to the token. In some other cases, a dialect, an accent, a locality of the NL input affects the meaning of the token. Additionally, there can be multiple valid choices corresponding to a token but only one of them correct based on the factors involved.
- the factors contemplated by the illustrative embodiments are related to the phonetic variations of a token as described herein. As such, the factors contemplated by the illustrative embodiments, which affect machine-understanding of NL tokens, are distinct from misspelling and typographical errors-type of reasons that affect correct token recognition. Presently, techniques exist to help an NLP machine to select the correct choice when misspelled tokens are encountered in textual NL inputs.
- misspelled tokens are mapped to the same correct word, e.g., misspellings such as “tirminate”, “termate”, and “termenate” are mapped to the correct selection - "terminate” - to assist the NLP machine to make the correct selection when a misspelled token is encountered.
- Some presently used NLP algorithms build large caches of misspellings mapped to correct spellings. Such caches can be large, but they are still far from exhaustive. For example, just for the English language cache, a single eight-character word can theoretically have 26 8 (208,827,064,576) possible variations. Some algorithms in this class of algorithms optimize the cache, e.g., by including only the most common misspellings. Still, the cache of mappings remains far from complete, is not scalable, and handles only a limited type of issues - the misspellings in textual inputs.
- Fuzzy matching is another class of algorithms used to map an NL token to a choice or selection from a set of selections.
- a fuzzy matching algorithm is a string matching algorithm that uses variations of edit distance algorithms as a means of finding similarities between a given token string from textual input and an available selection string in a set of selections. Fuzzy matching algorithms also operate on textual NL inputs, and are presently configured for correctly understanding misspelled character strings.
- fuzzy matching algorithms are designed to have a high recall at the cost of sacrificing precision.
- Recall is a fraction of relevant instances that are retrieved
- precision is the fraction of retrieved instances that are relevant.
- Precision can be seen as a measure of exactness or quality, whereas recall is a measure of completeness or quantity.
- Maximum precision indicates no false positives, and maximum recall indicates no false negatives.
- An embodiment includes a method that extracts a token from a Natural Language (NL) input.
- the embodiment computes a phonetic pattern corresponding to the token, the phonetic pattern comprising a sound pattern that represents a part of the token when the token is spoken.
- the embodiment creates new data from data of the phonetic pattern, the new data comprising a syllable sequence corresponding to the phonetic pattern.
- the embodiment changes a state of a data storage device by storing the new data in a matrix of syllable sequences corresponding to the token.
- the embodiment selects, by executing a fuzzy matching algorithm using a processor and a memory, an option that corresponds to the token, wherein the selecting is based on a syllable sequence in the matrix.
- the embodiment causes an improvement in the precision of the fuzzy matching algorithm by selecting a matching option for the NL token where the option is selected because the option also corresponds to a syllable sequence representative of the token.
- Another embodiment further computes a second syllable sequence corresponding to the phonetic pattern.
- the embodiment saves the second syllable sequence in the matrix.
- the embodiment enables configuring a plurality of syllable sequences that is representative of the token.
- Another embodiment further determines whether a stored phonetic pattern in a phonetic repository corresponds to the token.
- the embodiment selects, responsive to the stored phonetic pattern corresponding to the token, the stored phonetic pattern as the phonetic pattern.
- the embodiment uses a historically learned phonetic pattern from a previous occurrence of the token.
- Another embodiment further modifies, to form the phonetic pattern, a stored phonetic pattern corresponding to a second token in a phonetic repository, wherein the second token is comparable to the token by having a greater than a threshold degree of structural similarity with the token.
- the embodiment uses a phonetic pattern of a comparable but different token to construct a phonetic pattern of the token.
- the structural similarity exists because at least a threshold degree of similarity exists between a spelling of the token and a spelling of the second token.
- the embodiment provides one manner of selecting the comparable token.
- the structural similarity exists because at least a threshold degree of similarity exists between a length of a spelling of the token and a length of a spelling of the second token.
- the embodiment provides another manner of selecting the comparable token.
- the structural similarity exists because at least a threshold degree of similarity exists between a sequence of vowels in the token and a sequence of vowels in the second token.
- the embodiment provides another manner of selecting the comparable token.
- the embodiment provides another manner of selecting the comparable token.
- the sound pattern comprises a phonetic representation, wherein the structural similarity exists because at least a threshold degree of similarity exists between a number of phonetic
- the embodiment provides another manner of selecting the comparable token.
- the sound pattern comprises a phonetic representation, wherein the structural similarity exists because at least a threshold degree of similarity exists between an order of phonetic
- the embodiment provides another manner of selecting the comparable token.
- the sound pattern comprises a phonetic representation, wherein the structural similarity exists because at least a threshold degree of similarity exists between an emphasis on the phonetic representation in the token and an emphasis on the phonetic representation in the second token.
- the embodiment provides another manner of selecting the comparable token.
- Another embodiment further computes the phonetic pattern from a second phonetic pattern by adding a second sound pattern to the second phonetic pattern.
- the embodiment provides a manner of modifying the phonetic pattern of the comparable token to create the phonetic pattern of the token.
- Another embodiment further computes the phonetic pattern from a second phonetic pattern by removing a second sound pattern from the second phonetic pattern.
- the embodiment provides another manner of modifying the phonetic pattern of the comparable token to create the phonetic pattern of the token.
- Another embodiment further computes the phonetic pattern from a second phonetic pattern by replacing a second sound pattern in the second phonetic pattern with the sound pattern.
- the embodiment provides another manner of modifying the phonetic pattern of the comparable token to create the phonetic pattern of the token.
- Another embodiment further computes the phonetic pattern from a second phonetic pattern by changing a sequence of sound patterns in the second phonetic pattern.
- the embodiment provides another manner of modifying the phonetic pattern of the comparable token to create the phonetic pattern of the token.
- Another embodiment further computes the phonetic pattern from a second phonetic pattern by changing a duration of a second sound pattern in the second phonetic pattern.
- the embodiment provides another manner of modifying the phonetic pattern of the comparable token to create the phonetic pattern of the token.
- Another embodiment further computes the phonetic pattern from a second phonetic pattern by changing an emphasis on a second sound pattern in the second phonetic pattern.
- the embodiment provides another manner of modifying the phonetic pattern of the comparable token to create the phonetic pattern of the token.
- Another embodiment further computes a set of phonetic patterns corresponding to the token, the phonetic pattern being a member of the set of phonetic patterns.
- the embodiment provides that a plurality of phonetic patterns can represent the token.
- the sound pattern represents the part of the token when the token is spoken in a dialect of a language.
- a plurality of phonetic patterns can represent the token such that a phonetic pattern is based on sound patterns created by speaking in a dialect.
- the sound pattern represents the part of the token when the token is spoken with an accent in a language.
- a plurality of phonetic patterns can represent the token such that a phonetic pattern is based on sound patterns created by speaking with an accent.
- the sound pattern represents the part of the token when the token is spoken in a language with a speech-peculiarity of a speaker.
- a plurality of phonetic patterns can represent the token such that a phonetic pattern is based on sound patterns created by a speaker's speaking-related idiosyncrasies.
- the token comprises a shortest meaningful unit of speech in the NL input.
- the embodiment provides a specific type of linguistic construct that can form a token.
- An embodiment includes a computer usable program product.
- the computer usable program product includes a computer-readable storage device, and program instructions stored on the storage device.
- An embodiment includes a computer system.
- the computer system includes a processor, a computer-readable memory, and a computer-readable storage device, and program instructions stored on the storage device for execution by the processor via the memory.
- the illustrative embodiments recognize that factors other than misspellings in textual inputs are responsible for precision of understanding NL tokens. Such factors are dependent upon the tonal or phonetic characteristics of the token rather than the correctness or incorrectness of the textual spelling of the token. [0035]
- the illustrative embodiments recognize that a method is needed by which the phonetic variations of tokens can be represented in NLP so that the fuzzy matching application increase in precision while keeping the recall characteristic unchanged when making selections corresponding to NL inputs.
- the illustrative embodiments recognize that the presently available fuzzy matching algorithms have to be modified to be able to use phonetic characteristics of tokens as additional inputs in determining the correct selection corresponding to the token.
- Figure 1 depicts a block diagram of a network of data processing systems in which illustrative embodiments may be implemented
- Figure 2 depicts a block diagram of a data processing system in which illustrative embodiments may be implemented
- Figure 3 depicts a block diagram of an example configuration for phonetic patterns for fuzzy matching in natural language processing in accordance with an illustrative embodiment
- Figure 4 depicts a block diagram of an example application for phonetic patterns for fuzzy matching in natural language processing in accordance with an illustrative embodiment
- Figure 5 depicts a flowchart of an example process for phonetic patterns for fuzzy matching in natural language processing in accordance with an illustrative embodiment
- Figure 6 depicts a flowchart of an example process for machine learning to improve phonetic pattern selection for NL tokens in accordance with an illustrative embodiment.
- the illustrative embodiments recognize that the presently available tools or solutions do not address these needs/problems or provide adequate solutions for these needs/problems.
- the illustrative embodiments used to describe the invention generally address and solve the above-described problems and other related problems by phonetic patterns for fuzzy matching in natural language processing.
- An embodiment can be implemented as a software application.
- the application implementing an embodiment, or one or more components thereof can be configured as a modification of an existing application implementing an fuzzy matching algorithm (fuzzy matching application) - i.e., a native application in the fuzzy matching application, as an application executing in a data processing system communicating with an existing fuzzy matching application over a short-range radio network such as Bluetooth, over a wired or wireline local area network (LAN) - i.e., a local application on the LAN, as an application executing in a data processing system communicating with an existing fuzzy matching application over a wide area network (WAN) - i.e., a remote application on the WAN, as a separate application that operates in conjunction with an existing fuzzy matching application in other ways, a standalone application, or some combination thereof.
- fuzzy matching application i.e., a native application in the fuzzy matching application
- LAN local area network
- WAN wide area network
- WAN wide area network
- a phonetic representation is a sound pattern that represents all or a part of a NL token when the token is spoken.
- a syllable is phonological building block, to wit, a sound used in a speech in a given language that can be represented using one or more vowels of that language.
- a token is extracted from an NL input.
- the embodiment determines one or more distinct phonetic representations (distinct phonetic sounds) that together from a phonetic pattern, and the phonetic pattern represents the token.
- one manner of representing an example token - "tomato” - is where "tow” "mah” and “tow” form a set of three phonetic representations that together form the phonetic pattern “tow-mah-tow”, which represents token "tomato”.
- the embodiment constructs one or more phonetic patterns using corresponding sets of phonetic representations, such that any of the phonetic patterns can possibly represent the given token.
- An embodiment stores a phonetic pattern of a token in a repository - the phonetic repository. Over time, upon encountering several tokens and several instances of the same token under different circumstances, the phonetic repository becomes populated with one or more phonetic patterns corresponding to one or more tokens.
- a set of phonetic patterns corresponding to the token are available in the phonetic repository.
- the phonetic pattern is extracted into the set of phonetic patterns for that token.
- a phonetic pattern of a different token which is comparable or similar to the given token in its phonetic structure, may be available in the repository.
- An embodiment computes a phonetic pattern of the token from the phonetic pattern of the comparable token.
- a phonetic pattern for tomato may not be available but a phonetic pattern for "potato” might be available in the repository.
- An embodiment determines the structural similarity between a given token and a comparable token by comparing for the two tokens the spelling, the length of the spelling, arrangement of the vowels therein, the number of phonetic representations therein, the order of phonetic representations therein, the emphasis on a particular phonetic representation, or some combination of these and other such characteristics.
- the embodiment uses the phonetic pattern of the comparable token from the repository and computes a phonetic pattern of the given token from that phonetic pattern.
- the computing of the phonetic pattern of the given token can include adding a phonetic representation to the phonetic pattern of the comparable token, deleting a phonetic representation from the phonetic pattern of the comparable token, replacing a phonetic representation in the phonetic pattern of the comparable token with a different phonetic representation, changing a sequence of phonetic representations in the phonetic pattern of the comparable token, compressing a duration during which the phonetic pattern of the comparable token is spoken or sounded out, expanding a duration during which the phonetic pattern of the comparable token is spoken or sounded out, adding an emphasis to a phonetic representation in the phonetic pattern of the comparable token, removing an emphasis from a phonetic representation in the phonetic pattern of the comparable token, changing an emphasis from one phonetic representation to a different phonetic representation in the phonetic pattern of the comparable token, and a combination of these and many other aspects depending upon the language, dialect, accent, context, speaker's peculiarities, etc.
- an embodiment can construct a set of phonetic patterns corresponding to a token using phonetic patterns for the token from the repository and phonetic patterns of comparable tokens from the repositories.
- a phonetic pattern for a token can also be constructed by segmenting the audio of the token into constituent phonetic representations.
- an embodiment computes a syllable corresponding to each phonetic representation in the phonetic pattern.
- the embodiment transforms a phonetic pattern into a sequence of syllables.
- the transformation of a phonetic representation into a syllable can be performed by matching a sound present in the phonetic representation with a sound made by sounding out a syllable.
- the phonetic pattern "tow-mah-tow” includes the sequence of sounds "oh” (D), "ah” (a:), and "oh” (D).
- the sequence of syllables corresponding to the phonetic pattern of the token is "D a: D”.
- the embodiment constructs a syllable sequence for each phonetic pattern corresponding to the token. More than one syllable sequences may be possible for a single phonetic pattern.
- the set of syllable sequence thus created forms a matrix of syllable sequences.
- An embodiment modifies an existing fuzzy matching application to accept the matrix of syllable sequences as a supporting input together with the NL input.
- the modified fuzzy matching application uses a syllable sequence in the matrix in the fuzzy matching algorithm to identify a suitable selection corresponding to a token.
- the selections identified using the syllable sequences exhibit a higher precision than the selections identified using only the prior-art fuzzy matching algorithm without using the syllable sequences.
- a method of an embodiment described herein when implemented to execute on a device or data processing system, comprises substantial advancement of the functionality of that device or data processing system in improving a precision of a fuzzy matching application without disturbing the fuzzy matching application's recall characteristics in the technical field of NLP.
- the illustrative embodiments are described with respect to certain types of NL, NL inputs, grammars, tokens, phonetic representations, phonetic patterns, syllables, syllable sequences, matrices, fuzzy matching algorithms, fuzzy matching applications, devices, data processing systems, environments, components, and applications only as examples. Any specific manifestations of these and other similar artifacts are not intended to be limiting to the invention. Any suitable manifestation of these and other similar artifacts can be selected within the scope of the illustrative embodiments.
- the illustrative embodiments may be implemented with respect to any type of data, data source, or access to a data source over a data network.
- Any type of data storage device may provide the data to an embodiment of the invention, either locally at a data processing system or over a data network, within the scope of the invention.
- any type of data storage device suitable for use with the mobile device may provide the data to such embodiment, either locally at the mobile device or over a data network, within the scope of the illustrative embodiments.
- the illustrative embodiments are described using specific code, designs, architectures, protocols, layouts, schematics, and tools only as examples and are not limiting to the illustrative embodiments. Furthermore, the illustrative embodiments are described in some instances using particular software, tools, and data processing environments only as an example for the clarity of the description. The illustrative embodiments may be used in conjunction with other comparable or similarly purposed structures, systems, applications, or architectures. For example, other comparable mobile devices, structures, systems, applications, or architectures therefor, may be used in conjunction with such embodiment of the invention within the scope of the invention. An illustrative embodiment may be implemented in hardware, software, or a combination thereof.
- Figures 1 and 2 are only examples and are not intended to assert or imply any limitation with regard to the environments in which different embodiments may be implemented. A particular implementation may make many modifications to the depicted environments based on the following description.
- FIG. 1 depicts a block diagram of a network of data processing systems in which illustrative embodiments may be implemented.
- Data processing environment 100 is a network of computers in which the illustrative embodiments may be implemented.
- Data processing environment 100 includes network 102.
- Network 102 is the medium used to provide communications links between various devices and computers connected together within data processing environment 100.
- Network 102 may include connections, such as wire, wireless communication links, or fiber optic cables.
- Clients or servers are only example roles of certain data processing systems connected to network 102 and are not intended to exclude other configurations or roles for these data processing systems.
- Server 104 and server 106 couple to network 102 along with storage unit 108.
- Software applications may execute on any computer in data processing environment 100.
- Clients 110, 112, and 114 are also coupled to network 102.
- a data processing system, such as server 104 or 106, or client 110, 112, or 114 may contain data and may have software applications or software tools executing thereon.
- Figure 1 depicts certain components that are usable in an example implementation of an embodiment.
- servers 104 and 106, and clients 110, 112, 114 are depicted as servers and clients only as examples and not to imply a limitation to a client-server architecture.
- an embodiment can be distributed across several data processing systems and a data network as shown, whereas another embodiment can be implemented on a single data processing system within the scope of the illustrative embodiments.
- Data processing systems 104, 106, 110, 112, and 114 also represent example nodes in a cluster, partitions, and other configurations suitable for implementing an embodiment.
- Device 132 is an example of a device described herein.
- device 132 can take the form of a smartphone, a tablet computer, a laptop computer, client 110 in a stationary or a portable form, a wearable computing device, or any other suitable device.
- Any software application described as executing in another data processing system in Figure 1 can be configured to execute in device 132 in a similar manner.
- Any data or information stored or produced in another data processing system in Figure 1 can be configured to be stored or produced in device 132 in a similar manner.
- fuzzy matching application 107 is a modified fuzzy matching application which includes a modified fuzzy matching algorithm where the modified fuzzy matching algorithm has been modified to accept a matrix of syllable sequences as an additional input for identifying a correct selection corresponding to a token in an NL input.
- Phonetic repository 109 includes one or more phonetic patterns for one or more tokens in one or more languages, as described herein.
- Servers 104 and 106, storage unit 108, and clients 110, 112, and 114, and device 132 may couple to network 102 using wired connections, wireless communication protocols, or other suitable data connectivity.
- Clients 110, 112, and 114 may be, for example, personal computers or network computers.
- server 104 may provide data, such as boot files, operating system images, and applications to clients 110, 112, and 114.
- Clients 110, 112, and 114 may be clients to server 104 in this example.
- Clients 110, 112, 114, or some combination thereof, may include their own data, boot files, operating system images, and applications.
- Data processing environment 100 may include additional servers, clients, and other devices that are not shown.
- data processing environment 100 may be the Internet.
- Network 102 may represent a collection of networks and gateways that use the Transmission Control Protocol/Internet Protocol (TCP/IP) and other protocols to communicate with one another.
- TCP/IP Transmission Control Protocol/Internet Protocol
- At the heart of the Internet is a backbone of data communication links between major nodes or host computers, including thousands of commercial, governmental, educational, and other computer systems that route data and messages.
- data processing environment 100 also may be implemented as a number of different types of networks, such as for example, an intranet, a local area network (LAN), or a wide area network (WAN).
- Figure 1 is intended as an example, and not as an architectural limitation for the different illustrative embodiments.
- data processing environment 100 may be used for implementing a client-server environment in which the illustrative embodiments may be implemented.
- a client-server environment enables software applications and data to be distributed across a network such that an application functions by using the interactivity between a client data processing system and a server data processing system.
- Data processing environment 100 may also employ a service oriented architecture where interoperable software components distributed across a network may be packaged together as coherent business applications.
- Data processing environment 100 may also take the form of a cloud, and employ a cloud computing model of service delivery for enabling convenient, on-demand network access to a shared pool of configurable computing resources (e.g. networks, network bandwidth, servers, processing, memory, storage, applications, virtual machines, and services) that can be rapidly provisioned and released with minimal management effort or interaction with a provider of the service.
- configurable computing resources e.g. networks, network bandwidth, servers, processing, memory, storage, applications, virtual machines, and services
- Data processing system 200 is an example of a computer, such as servers 104 and 106, or clients 110, 112, and 114 in Figure 1, or another type of device in which computer usable program code or instructions implementing the processes may be located for the illustrative embodiments.
- Data processing system 200 is also representative of a data processing system or a configuration therein, such as data processing system 132 in Figure 1 in which computer usable program code or instructions implementing the processes of the illustrative embodiments may be located.
- Data processing system 200 is described as a computer only as an example, without being limited thereto. Implementations in the form of other devices, such as device 132 in Figure 1, may modify data processing system 200, such as by adding a touch interface, and even eliminate certain depicted components from data processing system 200 without departing from the general description of the operations and functions of data processing system 200 described herein.
- data processing system 200 employs a hub architecture including North Bridge and memory controller hub (NB/MCH) 202 and South Bridge and input/output (I/O) controller hub (SB/ICH) 204.
- N/MCH North Bridge and memory controller hub
- SB/ICH South Bridge and input/output controller hub
- Processing unit 206, main memory 208, and graphics processor 210 are coupled to North Bridge and memory controller hub (NB/MCH) 202.
- Processing unit 206 may contain one or more processors and may be implemented using one or more heterogeneous processor systems.
- Processing unit 206 may be a multi-core processor.
- Graphics processor 210 may be coupled to NB/MCH 202 through an accelerated graphics port (AGP) in certain implementations.
- AGP accelerated graphics port
- local area network (LAN) adapter 212 is coupled to South Bridge and I/O controller hub (SB/ICH) 204.
- Audio adapter 216, keyboard and mouse adapter 220, modem 222, read only memory (ROM) 224, universal serial bus (USB) and other ports 232, and PCI/PCIe devices 234 are coupled to South Bridge and I/O controller hub 204 through bus 238.
- Hard disk drive (HDD) or solid-state drive (SSD) 226 and CD-ROM 230 are coupled to South Bridge and I/O controller hub 204 through bus 240.
- PCI/PCIe devices 234 may include, for example, Ethernet adapters, add-in cards, and PC cards for notebook computers.
- ROM 224 may be, for example, a flash binary input/output system (BIOS).
- BIOS binary input/output system
- Hard disk drive 226 and CD-ROM 230 may use, for example, an integrated drive electronics (IDE), serial advanced technology attachment (SATA) interface, or variants such as external-SATA (eSATA) and micro- SATA (mSATA).
- IDE integrated drive electronics
- SATA serial advanced technology attachment
- eSATA external-SATA
- mSATA micro- SATA
- a super I/O (SIO) device 236 may be coupled to South Bridge and I/O controller hub (SB/ICH) 204 through bus 238.
- SIO South Bridge and I/O controller hub
- Memories such as main memory 208, ROM 224, or flash memory (not shown), are some examples of computer usable storage devices.
- Hard disk drive or solid state drive 226, CD-ROM 230, and other similarly usable devices are some examples of computer usable storage devices including a computer usable storage medium.
- An operating system runs on processing unit 206.
- the operating system coordinates and provides control of various components within data processing system 200 in Figure 2.
- the operating system may be a commercially available operating system for any type of computing platform, including but not limited to server systems, personal computers, and mobile devices.
- An object oriented or other type of programming system may operate in conjunction with the operating system and provide calls to the operating system from programs or applications executing on data processing system 200.
- Instructions for the operating system, the object-oriented programming system, and applications or programs, such as application 105 and/or modified fuzzy matching application 107 in Figure 1, are located on storage devices, such as in the form of code 226A on hard disk drive 226, and may be loaded into at least one of one or more memories, such as main memory 208, for execution by processing unit 206.
- the processes of the illustrative embodiments may be performed by processing unit 206 using computer implemented instructions, which may be located in a memory, such as, for example, main memory 208, read only memory 224, or in one or more peripheral devices.
- code 226A may be downloaded over network 201 A from remote system 201 B, where similar code 201C is stored on a storage device 201 D. in another case, code 226A may be downloaded over network 201 A to remote system 201 B, where downloaded code 201C is stored on a storage device 201 D.
- the hardware in Figures 1-2 may vary depending on the implementation.
- Other internal hardware or peripheral devices such as flash memory, equivalent non-volatile memory, or optical disk drives and the like, may be used in addition to or in place of the hardware depicted in Figures 1-2.
- the processes of the illustrative embodiments may be applied to a multiprocessor data processing system.
- data processing system 200 may be a personal digital assistant (PDA), which is generally configured with flash memory to provide non-volatile memory for storing operating system files and/or user-generated data.
- PDA personal digital assistant
- a bus system may comprise one or more buses, such as a system bus, an I/O bus, and a PCI bus.
- the bus system may be implemented using any type of communications fabric or architecture that provides for a transfer of data between different components or devices attached to the fabric or architecture.
- a communications unit may include one or more devices used to transmit and receive data, such as a modem or a network adapter.
- a memory may be, for example, main memory 208 or a cache, such as the cache found in North Bridge and memory controller hub 202.
- a processing unit may include one or more processors or CPUs.
- data processing system 200 also may be a tablet computer, laptop computer, or telephone device in addition to taking the form of a mobile or wearable device.
- a computer or data processing system is described as a virtual machine, a virtual device, or a virtual component
- the virtual machine, virtual device, or the virtual component operates in the manner of data processing system 200 using virtualized manifestation of some or all components depicted in data processing system 200.
- processing unit 206 is manifested as a virtualized instance of all or some number of hardware processing units 206 available in a host data processing system
- main memory 208 is manifested as a virtualized instance of all or some portion of main memory 208 that may be available in the host data processing system
- disk 226 is manifested as a virtualized instance of all or some portion of disk 226 that may be available in the host data processing system.
- the host data processing system in such cases is represented by data processing system 200.
- this figure depicts a block diagram of an example configuration for phonetic patterns for fuzzy matching in natural language processing in accordance with an illustrative embodiment.
- Application 302 is an example of application 105 in Figure 1.
- Fuzzy matching application 304 is an example of modified fuzzy matching application 107 in Figure 1.
- Fuzzy matching application 304 comprises prior-art fuzzy matching algorithm 304A and phonetic input processing component 304B according to an embodiment.
- Phonetic repository 306 is an example of phonetic repository 109 in Figure 1.
- NL input 308 comprises one or more forms of NL data as described herein.
- NL input 308 includes one or more tokens, e.g., words.
- Application 302 suitably parses NL input 308 to extract a token from NL input 308. Using one or more phonetic patterns from phonetic repository 306 and/or audio data from NL input 308 if available, application 302 constructs one or more phonetic patterns corresponding to the token. Application 302 computes one or more syllable sequences corresponding to each phonetic pattern and forms syllable sequence matrix 310.
- a syllable sequence, e.g., syllable sequence 312 in matrix 310 comprises a sequence of one or more syllables, such as syllables 312A, 312B...312C.
- Phonetic input processing component 304B receives NL input 308 as one input, and matrix 310 from application 302 as an additional input.
- Component 304B causes fuzzy matching algorithm 304A to use the syllable sequences in the selection/choice identification process corresponding to a given token.
- Fuzzy matching application 304 outputs selection 314 with a greater precision and at least the same recall as compared to a selection that would have been output from fuzzy matching algorithm 304A alone.
- the selection output of modified fuzzy matching application 304 is used to train application 302. For example, if a separate process (not shown) determines that selection 314 is a probable selection for a given token, machine learning feedback 316 causes application 302 to increase or reinforce those phonetic patterns that correspond to selection 314 within a threshold degree of correspondence such that those phonetic patterns are produced again (or given comparatively more weight) for the same or similar token in a subsequent occurrence of the token. If a phonetic pattern was constructed that does not correspond to selection 314 within a threshold degree of correspondence, feedback 316 causes application 302 to decrease or weaken those phonetic patterns such that those phonetic patterns are not produced (or given comparatively less weight) for the same or similar token in a subsequent occurrence of the token.
- this figure depicts a block diagram of an example application for phonetic patterns for fuzzy matching in natural language processing in accordance with an illustrative embodiment.
- Application 402 can be used as application 302 in Figure 3.
- Component 404 extracts a token from an NL input, e.g., by parsing NL input 308 according to a grammar of the language of input 308.
- Component 406 computes a phonetic pattern for the token in any one or more manners described herein.
- Component 408 computes a syllable sequence corresponding to a phonetic pattern, component 408 outputs a matrix of syllable sequences, e.g., matrix 310, corresponding to the token.
- Component 410 may be configured to operate as phonetic input processing component 304B in Figure 3.
- Component 410 processes the matrix of syllable sequences to use in the modified fuzzy matching application, e.g., in fuzzy matching application 304.
- Component 412 receives a feedback of the selection made by the modified fuzzy matching application, e.g., feedback 316.
- Component 414 implements a suitable process for classifying the selection as probable (valid selection in the context of NL input 308) or improbable (invalid selection in the context of NL input 308).
- Component 416 reinforces those phonetic patterns of the token which correspond to a probable selection.
- Component 418 weakens those phonetic patterns of the token which correspond to an improbable selection.
- Component 420 manages the phonetic repository, e.g., by adding a new phonetic pattern, reinforcing a phonetic pattern, weakening a phonetic pattern, removing a phonetic pattern, or otherwise manipulating phonetic patterns in the repository.
- FIG. 5 depicts a flowchart of an example process for phonetic patterns for fuzzy matching in natural language processing in accordance with an illustrative embodiment.
- Process 500 can be implemented in application 402 in Figure 4.
- the application receives an NL input (block 502).
- the application selects a unit of speech, to wit, a token, from the input (block 504).
- the application performs one or more of blocks 506, 508, and 510.
- the application determines whether a phonetic pattern for the token exists in a phonetic repository (block 506); the application determines whether a phonetic pattern of a comparable token exists in the repository (block 508); and/or the application computes a phonetic pattern of the token based on the actual or estimated audio of the token in the NL input (block 510). If/when the application executes block 510 to compute a phonetic pattern of the token based on the actual or estimated audio of the token in the NL input, the application updates the phonetic repository with the computed phonetic pattern (block 512).
- the application computes one or more syllable sequences corresponding to the phonetic pattern (block 514). If a phonetic pattern of the token does not exist in the repository ("No” path of block 506), the application proceeds to block 508.
- the application computes a phonetic pattern of the token based on the phonetic pattern of the comparable token (block 516).
- the application updates the repository with the computed phonetic pattern for the token at block 512.
- the application determines whether more phonetic patterns of the token have to be computed in a similar manner (block 518). If more phonetic patterns of the token have to be computed ("Yes” path of block 518), the application returns to any of blocks 506, 508, and 510. If no more phonetic patterns have to be computed for the token ("No” path of block 518), the application generates a matrix of the computed syllable sequences (block 520). Optionally, the application may preprocess the matrix for consumption as an additional input into a modified fuzzy matching application (block 522).
- the application provides the matrix as an additional input to the modified fuzzy matching application (block 524).
- the application causes the modified fuzzy matching application to output a selection corresponding to the token with an improved precision (block 526).
- the application may end process 500 thereafter.
- the application may also determine in parallel whether more tokens have to be processed from the NL input (block 528). If more tokens have to be processed from the NL input ("Yes” path of block 528), the application returns to block 504 and selects another token. If no more tokens have to be processed from the NL input ("No” path of block 528), the application ends process 500 thereafter.
- FIG. 6 depicts a flowchart of an example process for machine learning to improve phonetic pattern selection for NL tokens in accordance with an illustrative embodiment.
- Process 600 can be implemented in application 402 in Figure 4.
- the application receives a selection made by a modified fuzzy matching application corresponding to a token (block 602).
- the application determines whether, given the NL input context, the selection is probable (block 604). If the selection is probable ("Probable” path of block 604), the application reinforces those phonetic patterns of the token which correspond to the selection (block 606). The application ends process 600 thereafter.
- the application weakens those phonetic patterns of the token which correspond to the selection (block 608). The application ends process 600 thereafter.
- a computer implemented method, system or apparatus, and computer program product are provided in the illustrative embodiments for phonetic patterns for fuzzy matching in natural language processing and other related features, functions, or operations. Where an embodiment or a portion thereof is described with respect to a type of device, the computer implemented method, system or apparatus, the computer program product, or a portion thereof, are adapted or configured for use with a suitable and comparable manifestation of that type of device.
- SaaS Software as a Service
- a SaaS model the capability of the application implementing an embodiment is provided to a user by executing the application in a cloud infrastructure.
- the user can access the application using a variety of client devices through a thin client interface such as a web browser (e.g., web-based e-mail), or other light-weight client-applications.
- the user does not manage or control the underlying cloud infrastructure including the network, servers, operating systems, or the storage of the cloud infrastructure.
- the user may not even manage or control the capabilities of the SaaS application.
- the SaaS implementation of the application may permit a possible exception of limited user-specific application configuration settings.
- the present invention may be a system, a method, and/or a computer program product at any possible technical detail level of integration
- the computer program product may include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the present invention
- the computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device.
- the computer readable storage medium may be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing.
- a non-exhaustive list of more specific examples of the computer readable storage medium includes the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanically encoded device such as punch-cards or raised structures in a groove having instructions recorded thereon, and any suitable combination of the foregoing.
- RAM random access memory
- ROM read-only memory
- EPROM or Flash memory erasable programmable read-only memory
- SRAM static random access memory
- CD-ROM compact disc read-only memory
- DVD digital versatile disk
- memory stick a floppy disk
- a mechanically encoded device such as punch-cards or raised structures in a groove having instructions recorded thereon
- a computer readable storage medium including but not limited to computer-readable storage devices as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.
- Computer readable program instructions described herein can be downloaded to respective computing/processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and/or a wireless network.
- the network may comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers.
- a network adapter card or network interface in each computing/processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing/processing device.
- Computer readable program instructions for carrying out operations of the present invention may be assembler instructions, instruction-set-architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state-setting data, configuration data for integrated circuitry, or either source code or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C++, or the like, and procedural programming languages, such as the "C" programming language or similar programming languages.
- the computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server.
- the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider).
- electronic circuitry including, for example, programmable logic circuitry, field-programmable gate arrays (FPGA), or programmable logic arrays (PLA) may execute the computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry, in order to perform aspects of the present invention.
- These computer readable program instructions may be provided to a processor of a general-purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks.
- These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, a programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable storage medium having instructions stored therein comprises an article of manufacture including instructions which implement aspects of the function/act specified in the flowchart and/or block diagram block or blocks.
- the computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process, such that the instructions which execute on the computer, other programmable apparatus, or other device implement the functions/acts specified in the flowchart and/or block diagram block or blocks.
- each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s).
- the functions noted in the blocks may occur out of the order noted in the Figures.
- two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Acoustics & Sound (AREA)
- Human Computer Interaction (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Machine Translation (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Document Processing Apparatus (AREA)
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
GB2008024.8A GB2585492B (en) | 2017-11-15 | 2018-10-31 | Phonetic patterns for fuzzy matching in natural language processing |
JP2020524106A JP7135084B2 (ja) | 2017-11-15 | 2018-10-31 | 自然言語処理におけるファジー・マッチングのための音声パターン |
DE112018005844.1T DE112018005844B4 (de) | 2017-11-15 | 2018-10-31 | Verfahren, Programmprodukt, sowie Computersystem zum Fuzzy Matching von phonetischen Mustern bei der Verarbeitung von natürlicher Sprache |
CN201880073075.9A CN111328416B (zh) | 2017-11-15 | 2018-10-31 | 用于自然语言处理中的模糊匹配的语音模式 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US15/813,416 | 2017-11-15 | ||
US15/813,416 US10546062B2 (en) | 2017-11-15 | 2017-11-15 | Phonetic patterns for fuzzy matching in natural language processing |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2019097335A1 true WO2019097335A1 (en) | 2019-05-23 |
Family
ID=66433281
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/IB2018/058521 WO2019097335A1 (en) | 2017-11-15 | 2018-10-31 | Phonetic patterns for fuzzy matching in natural language processing |
Country Status (6)
Country | Link |
---|---|
US (2) | US10546062B2 (de) |
JP (1) | JP7135084B2 (de) |
CN (1) | CN111328416B (de) |
DE (1) | DE112018005844B4 (de) |
GB (1) | GB2585492B (de) |
WO (1) | WO2019097335A1 (de) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10546062B2 (en) * | 2017-11-15 | 2020-01-28 | International Business Machines Corporation | Phonetic patterns for fuzzy matching in natural language processing |
US11568142B2 (en) * | 2018-06-04 | 2023-01-31 | Infosys Limited | Extraction of tokens and relationship between tokens from documents to form an entity relationship map |
CN114927128A (zh) * | 2022-04-21 | 2022-08-19 | 清华大学 | 语音关键词的检测方法、装置、电子设备及可读存储介质 |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070055525A1 (en) * | 2005-08-31 | 2007-03-08 | Kennewick Robert A | Dynamic speech sharpening |
US8521539B1 (en) * | 2012-03-26 | 2013-08-27 | Nuance Communications, Inc. | Method for chinese point-of-interest search |
US20150340035A1 (en) * | 2014-02-07 | 2015-11-26 | Nexgen Flight, Llc. | Automated generation of phonemic lexicon for voice activated cockpit management systems |
Family Cites Families (37)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5794194A (en) * | 1989-11-28 | 1998-08-11 | Kabushiki Kaisha Toshiba | Word spotting in a variable noise level environment |
US6236964B1 (en) * | 1990-02-01 | 2001-05-22 | Canon Kabushiki Kaisha | Speech recognition apparatus and method for matching inputted speech and a word generated from stored referenced phoneme data |
JP2001100776A (ja) * | 1999-09-30 | 2001-04-13 | Arcadia:Kk | 音声合成装置 |
US8392188B1 (en) * | 1999-11-05 | 2013-03-05 | At&T Intellectual Property Ii, L.P. | Method and system for building a phonotactic model for domain independent speech recognition |
US7177795B1 (en) * | 1999-11-10 | 2007-02-13 | International Business Machines Corporation | Methods and apparatus for semantic unit based automatic indexing and searching in data archive systems |
US6973427B2 (en) * | 2000-12-26 | 2005-12-06 | Microsoft Corporation | Method for adding phonetic descriptions to a speech recognition lexicon |
US7103542B2 (en) * | 2001-12-14 | 2006-09-05 | Ben Franklin Patent Holding Llc | Automatically improving a voice recognition system |
US7089188B2 (en) * | 2002-03-27 | 2006-08-08 | Hewlett-Packard Development Company, L.P. | Method to expand inputs for word or document searching |
US20050182558A1 (en) * | 2002-04-12 | 2005-08-18 | Mitsubishi Denki Kabushiki Kaisha | Car navigation system and speech recognizing device therefor |
EP1505572B1 (de) * | 2002-05-06 | 2006-03-08 | Prous Institute For Biomedical Research S.A. | Spracherkennungsverfahren |
DE10306022B3 (de) * | 2003-02-13 | 2004-02-19 | Siemens Ag | Dreistufige Einzelworterkennung |
US20060064177A1 (en) * | 2004-09-17 | 2006-03-23 | Nokia Corporation | System and method for measuring confusion among words in an adaptive speech recognition system |
US8626506B2 (en) * | 2006-01-20 | 2014-01-07 | General Motors Llc | Method and system for dynamic nametag scoring |
WO2008084476A2 (en) | 2007-01-09 | 2008-07-17 | Avraham Shpigel | Vowel recognition system and method in speech to text applications |
JP4973731B2 (ja) * | 2007-07-09 | 2012-07-11 | 富士通株式会社 | 音声認識装置、音声認識方法、および、音声認識プログラム |
US8831946B2 (en) * | 2007-07-23 | 2014-09-09 | Nuance Communications, Inc. | Method and system of indexing speech data |
US9405823B2 (en) * | 2007-07-23 | 2016-08-02 | Nuance Communications, Inc. | Spoken document retrieval using multiple speech transcription indices |
US20090037403A1 (en) | 2007-07-31 | 2009-02-05 | Microsoft Corporation | Generalized location identification |
US8775441B2 (en) * | 2008-01-16 | 2014-07-08 | Ab Initio Technology Llc | Managing an archive for approximate string matching |
CN101533398A (zh) * | 2008-03-16 | 2009-09-16 | 徐文新 | 模式匹配索引查找方法 |
US20110106792A1 (en) | 2009-11-05 | 2011-05-05 | I2 Limited | System and method for word matching and indexing |
EP2509005A1 (de) * | 2009-12-04 | 2012-10-10 | Sony Corporation | Suchvorrichtung, suchverfahren und programm |
US8914385B2 (en) * | 2010-02-24 | 2014-12-16 | Mitsubishi Electric Corporation | Search device and search program |
US9418152B2 (en) * | 2011-02-09 | 2016-08-16 | Nice-Systems Ltd. | System and method for flexible speech to text search mechanism |
KR101231438B1 (ko) * | 2011-05-25 | 2013-02-07 | 엔에이치엔(주) | 외래어 발음 검색 서비스를 제공하는 검색결과 제공 시스템 및 방법 |
US9164983B2 (en) | 2011-05-27 | 2015-10-20 | Robert Bosch Gmbh | Broad-coverage normalization system for social media language |
US9721563B2 (en) | 2012-06-08 | 2017-08-01 | Apple Inc. | Name recognition system |
CN103970798B (zh) * | 2013-02-04 | 2019-05-28 | 商业对象软件有限公司 | 数据的搜索和匹配 |
US9652511B2 (en) | 2013-03-13 | 2017-05-16 | International Business Machines Corporation | Secure matching supporting fuzzy data |
US9594742B2 (en) * | 2013-09-05 | 2017-03-14 | Acxiom Corporation | Method and apparatus for matching misspellings caused by phonetic variations |
US9384731B2 (en) * | 2013-11-06 | 2016-07-05 | Microsoft Technology Licensing, Llc | Detecting speech input phrase confusion risk |
JP6400936B2 (ja) * | 2014-04-21 | 2018-10-03 | シノイースト・コンセプト・リミテッド | 音声検索方法、音声検索装置、並びに、音声検索装置用のプログラム |
CN105632499B (zh) * | 2014-10-31 | 2019-12-10 | 株式会社东芝 | 用于优化语音识别结果的方法和装置 |
CN104464736B (zh) | 2014-12-15 | 2018-02-02 | 北京百度网讯科技有限公司 | 语音识别文本的纠错方法和装置 |
GB2533370A (en) | 2014-12-18 | 2016-06-22 | Ibm | Orthographic error correction using phonetic transcription |
EP3353679A4 (de) * | 2015-09-22 | 2019-05-22 | Nuance Communications, Inc. | Systeme und verfahren zur erkennung von interessenspunkten |
US10546062B2 (en) * | 2017-11-15 | 2020-01-28 | International Business Machines Corporation | Phonetic patterns for fuzzy matching in natural language processing |
-
2017
- 2017-11-15 US US15/813,416 patent/US10546062B2/en active Active
-
2018
- 2018-10-31 CN CN201880073075.9A patent/CN111328416B/zh active Active
- 2018-10-31 JP JP2020524106A patent/JP7135084B2/ja active Active
- 2018-10-31 WO PCT/IB2018/058521 patent/WO2019097335A1/en active Application Filing
- 2018-10-31 DE DE112018005844.1T patent/DE112018005844B4/de active Active
- 2018-10-31 GB GB2008024.8A patent/GB2585492B/en active Active
-
2019
- 2019-11-26 US US16/696,098 patent/US11397856B2/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070055525A1 (en) * | 2005-08-31 | 2007-03-08 | Kennewick Robert A | Dynamic speech sharpening |
US8521539B1 (en) * | 2012-03-26 | 2013-08-27 | Nuance Communications, Inc. | Method for chinese point-of-interest search |
US20150340035A1 (en) * | 2014-02-07 | 2015-11-26 | Nexgen Flight, Llc. | Automated generation of phonemic lexicon for voice activated cockpit management systems |
Non-Patent Citations (1)
Title |
---|
JIANNAN WANG ET AL.: "Fast-Join: An Efficient Method for Fuzzy Token Matching based String Similarity Join", 2011 IEEE 27TH INTERNATIONAL CONFERENCE ON DATA ENGINEERING, 11 April 2011 (2011-04-11) - 31 December 2011 (2011-12-31), pages 458 - 469, XP031868459 * |
Also Published As
Publication number | Publication date |
---|---|
DE112018005844B4 (de) | 2021-11-25 |
JP7135084B2 (ja) | 2022-09-12 |
US20200097540A1 (en) | 2020-03-26 |
CN111328416B (zh) | 2023-11-17 |
GB2585492A (en) | 2021-01-13 |
CN111328416A (zh) | 2020-06-23 |
US10546062B2 (en) | 2020-01-28 |
US20190147036A1 (en) | 2019-05-16 |
GB2585492B (en) | 2021-03-17 |
GB202008024D0 (en) | 2020-07-15 |
US11397856B2 (en) | 2022-07-26 |
DE112018005844T5 (de) | 2020-08-13 |
JP2021503097A (ja) | 2021-02-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11769480B2 (en) | Method and apparatus for training model, method and apparatus for synthesizing speech, device and storage medium | |
US10832658B2 (en) | Quantized dialog language model for dialog systems | |
US10019438B2 (en) | External word embedding neural network language models | |
US10755048B2 (en) | Artificial intelligence based method and apparatus for segmenting sentence | |
JP2022534390A (ja) | ストリーミングエンドツーエンドモデルを用いる大規模多言語音声認識 | |
US11907671B2 (en) | Role labeling method, electronic device and storage medium | |
US11397856B2 (en) | Phonetic patterns for fuzzy matching in natural language processing | |
US10162812B2 (en) | Natural language processing system to analyze mobile application feedback | |
US11455335B2 (en) | Image retrieval using interactive natural language dialog | |
US11157707B2 (en) | Natural language response improvement in machine assisted agents | |
US11270686B2 (en) | Deep language and acoustic modeling convergence and cross training | |
US10592542B2 (en) | Document ranking by contextual vectors from natural language query | |
US11132507B2 (en) | Cross-subject model-generated training data for relation extraction modeling | |
US11238235B2 (en) | Automated novel concept extraction in natural language processing | |
US10565982B2 (en) | Training data optimization in a service computing system for voice enablement of applications | |
US10553203B2 (en) | Training data optimization for voice enablement of applications | |
US11710480B2 (en) | Phonetic comparison for virtual assistants | |
CN114444462B (zh) | 模型训练方法及人机交互方法、装置 | |
US10922486B2 (en) | Parse tree based vectorization for natural language processing | |
US10811009B2 (en) | Automatic skill routing in conversational computing frameworks | |
JP7349523B2 (ja) | 音声認識方法、音声認識装置、電子機器、記憶媒体コンピュータプログラム製品及びコンピュータプログラム | |
WO2023082831A1 (en) | Global neural transducer models leveraging sub-task networks | |
KR20230156795A (ko) | 단어 분할 규칙화 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 18879313 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2020524106 Country of ref document: JP Kind code of ref document: A |
|
ENP | Entry into the national phase |
Ref document number: 202008024 Country of ref document: GB Kind code of ref document: A Free format text: PCT FILING DATE = 20181031 |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 18879313 Country of ref document: EP Kind code of ref document: A1 |